Re: ASF board report draft for November 2024

2024-11-13 Thread Matei Zaharia
There’s still discussion about GraphX after the vote, so I will not that in the report instead of just saying we decided to deprecate it. But otherwise I’ll send the report as is. Thanks for the feedback everyone. > On Nov 12, 2024, at 11:52 AM, Bjørn Jørgensen > wrote: > > ohh.. my fault >

Re: [DISCUSS] Deprecate GraphX OR Find new maintainers interested in GraphX OR leave it as is?

2024-11-13 Thread Mich Talebzadeh
Hm. Since it sounds like a plan why Russell you go ahead and create a SPIP for it, then, this discussion takes a formal approach and is documented. Otherwise we are just flogging a dead horse so to speak. HTH Mich Talebzadeh, Architect | Data Engineer | Data Science | Financial Crime PhD

Re: [DISCUSS] Deprecate GraphX OR Find new maintainers interested in GraphX OR leave it as is?

2024-11-13 Thread Russell Jurney
It might be, but graph processing is a desirable, very useful feature of Spark. GraphX doesn't see more popularity because it never got a DataFrame interface. If someone is willing to add one and maintain it, that seems best of all. Russ On Wed, Nov 13, 2024 at 7:12 AM Ángel wrote: > Seems to m

Re: [DISCUSS] Deprecate GraphX OR Find new maintainers interested in GraphX OR leave it as is?

2024-11-13 Thread Holden Karau
Responses in line Twitter: https://twitter.com/holdenkarau Fight Health Insurance: https://www.fighthealthinsurance.com/ Books (Learning Spark, High Performance Spark, etc.): https://amzn.to/2MaRAG9 YouTube Live Streams:

Re: [DISCUSS] Deprecate GraphX OR Find new maintainers interested in GraphX OR leave it as is?

2024-11-13 Thread Denny Lee
> If by "at length" you mean ... among very few people, in a very few emails, throughout less than a week. Yes, you're totally right. The implication that this was a rash decision decided by a few people in a short amount of time is incorrect. This discussion all started on the thread "[VOTE] Off

Re: [DISCUSS] Deprecate GraphX OR Find new maintainers interested in GraphX OR leave it as is?

2024-11-13 Thread Ángel
If by "at length" you mean ... among very few people, in a very few emails, throughout less than a week. Yes, you're totally right. El mié, 13 nov 2024 a las 6:15, Sean Owen () escribió: > I think this was all discussed at length. > While Spark of course doesn't collect usage stats, we do have ev

Re: [DISCUSS] Deprecate GraphX OR Find new maintainers interested in GraphX OR leave it as is?

2024-11-13 Thread Sean Owen
I do; I do not work on Spark there. I do not see why it would affect support as the code is still part of Spark, but, it is off-topic for this list. On Wed, Nov 13, 2024 at 9:19 AM Ángel wrote: > Btw, Sean, you work for Databricks ... deprecating GraphX would mean ... > Databricks won't give sup

Re: [DISCUSS] Deprecate GraphX OR Find new maintainers interested in GraphX OR leave it as is?

2024-11-13 Thread Ángel
Btw, Sean, you work for Databricks ... deprecating GraphX would mean ... Databricks won't give support to this API anymore? for all versions supported or only for the new ones? I'm just curious about that. El mié, 13 nov 2024 a las 16:16, Ángel () escribió: > If by "at length" you mean ... among

Re: [DISCUSS] Deprecate GraphX OR Find new maintainers interested in GraphX OR leave it as is?

2024-11-13 Thread Ángel
Seems to me it would be easier to move GraphX to graphframes than the opposite. El mar, 8 oct 2024 a las 21:52, Reynold Xin () escribió: > We can also consider the following: move GraphFrame into Spark, and make > GraphX an internal impl detail of GraphFrame. Then we can over time change > th

Re: [DISCUSS] Support spark.ml on Spark Connect

2024-11-13 Thread Mich Talebzadeh
OK I added a comment to PR HTH, Mich Talebzadeh, Architect | Data Engineer | Data Science | Financial Crime PhD Imperial College London London, United Kingdom view my Linkedin profil

来自李天祥的邮件

2024-11-13 Thread 李天祥
退订

Re: [DISCUSS] Deprecate GraphX OR Find new maintainers interested in GraphX OR leave it as is?

2024-11-13 Thread Mich Talebzadeh
Insofar as I can ascertain, this topic has become one of the longest-running threads, with over 70 exchanges, and is now diverging from its initial focus. Given the length and tangential nature of the current discussion, I suggest setting a grace period of six months for volunteers to step forward

Re: [DISCUSS] Support spark.ml on Spark Connect

2024-11-13 Thread Bobby
Hi all, Finally, I got the draft PR up https://github.com/apache/spark/pull/48791 last week, but I would like to send it out here and hope to get your feedback and reviews. Thx very much. Mich Talebzadeh 于2024年10月15日周二 15:47写道: > +1 > > It will be a desirable feature > > Mich Talebzadeh, > > A